Dear Fellow White Men in Tech: Stop It.
I used to sit in class at Union Theological Seminary listening to Dr. Cornell West lecture. His words were prophetic poetry; a spoken word poem giving life to black liberation theology. Every class he would, without fail, give thanks to those in what he called his black ancestry: the heroes of jazz, Harriet Tubman, Jimmy Baldwin (as he would call him), the great giants of the blues tradition, the list would go on. For Dr. West there was no future liberation without paying tribute to the liberation fought for by those spiritual ancestors that came before. I would sit there in that class as a 22 year-old white man, just out of college, mesmerized by the passion and respect this powerful force had for all of those who had come before.
Recently my colleague, Dr. Timnit Gebru, tweeted a link to a timeline published by machinelearningknowledge.ai on November 24, 2019. This timeline outlined the “Brief History of Deep Learning from 1943–2019.” Dr. Gebru’s tweet read, “The male only history of deep learning, where you say AlexNet makes history but ImageNet doesn’t, because women’s contributions don’t count. And contributions from anyone except for white & white adjacent people for that matter.” I was curious about what Dr. Gebru was referring to and so I quickly clicked the link.
As an anthropologist of Artificial Intelligence ethics and someone whose research has revolved around the history of the development of deep learning, I like to think I have a good sense of what might be on such a timeline. But what I saw shocked me. On this timeline there was not a single woman acknowledged, only two non-white individuals acknowledged (one of which was a GO player and not a researcher), and no explanation for why.
Let us be clear about three points:
- Representation Matters.
Who we see represented alters our conception of our history, our present, and our future. Who we lift up as founding scholars in a field has immeasurable downstream impacts towards who will gravitate towards the field in the future and what questions the field will ask going forward.
2. Facts Matter.
This is not a case of a timeline simply not having good examples of non-white non-male scholars to draw from in the field. Dr. Fei-Fei Li altered the field of deep learning forever with her research leading to the creation of ImageNet. Dr. Daphne Koller made enormous strides in ensuring that students of all economic backgrounds could embark on a career grounded in deep learning in the first place. Dr. Latanya Sweeney has made incalculable contributions in the application of big data to deep learning and its use in privacy and security. The list goes on (and should be one day soon written in full!)
3. This is the Norm, Not the Exception.
The mono-identity representation within this timeline is not a unique telling of the history of deep learning. This essay is in no way a hit-piece on machinelearningknowledge.ai. Their timeline is, unfortunately, a microcosm for how industry and academia alike demonstrably default to valorizing the contributions of white men over the contributions of women and people of color. When I discuss the evolution of deep learning with colleagues at conferences, rarely do we begin the story of deep learning with the important contributions of women or people of color to the field, although there are many.
How we tell the story of our history is a moral statement, a political statement, and a philosophical statement. How we tell the story of the founding scholars of a field such as deep learning is a statement of what we value in the past, present, and future of the field. It is a factual mistake to ignore the contributions of women and people of color in the development of this field. However, it is also an ethical violation.
This is an essay for everyone, but especially for white men who are involved in telling the stories of the development of technology or who are developing technology themselves. I say the following as a white man in technology industry and within academia. It is time for us as white men to de-center our contributions in favor of the contributions of folks who have been historically marginalized in technology spaces. In no uncertain terms this means intentionally giving up institutional power, unmooring ourselves from our fixation on the spotlight, and most importantly, striving at every level to tell the stories of those who have come before — all of them.
I left Dr. West’s class more often than not trying to remember my own ancestors, the ones that had willed me into being. Exalting the memory of those who have gone before is a spiritual practice. We must challenge ourselves to tell the story of those who have come before fully. Further, because of the socio-political history we are living in the shadow of we must ensure our research actively and unapologetically centers the stories of those that the colonial march of time has tragically hidden.
The essay up to this point has been a flowery way to simply say: dear fellow white men in tech — stop it. Don’t be pissed. Don’t be defensive. Just stop erasing the stories in which you are not the protagonist.
